Our Central Hypothesis is that immunization of hepatocellular cancer (HCC) patients with a rationally-designed dendritic cell (DC)-based vaccine which presents HCC- associated tumor antigens AFP, glypican-3 and MAGE-A3 will result in activation of CD8+ and CD4+ T cells which recognize tumors. This broad, polyclonal, multi-antigen immunity will also activate NK cells and improve time to progression (TTP) and health-related quality of life (HRQL). We will test a DC vaccine differentially loaded with 3 defined antigens to promote broad immunity. DC will be protein-fed, adenovirally-transduced and peptide-pulsed. This will result in antigen presence in endosomes (from protein-fed antigen), in the cytoplasm (AdV-delivered antigen) and on the surface (peptide-pulsed). This setting will allow direct comparison of these three modes of antigen presentation from a single vaccine, and also allow for a comparison of vaccine-delivered CD4+ T cell help (AFP, GPC3) with direct CD8+ activation only (MAGE-A3) on the frequency changes, functional quality and longevity of the corresponding CD8+ T cells. A.1 Specific Aim 1. Antigen Processing and Presentation. In this aim, we will analyze the intracellular processing of the loaded antigens in HBV+/HCV+ HCC patient DC in a preclinical, in vitro, setting. We will address several mechanistic questions about the differentially loaded DC vaccines: mechanism of protein uptake; intracellular antigen localization; duration of antigen presence and presentation; the effect of DC maturation; and the signal transduction pathways activated. A.2 Specific Aim 2. Lymphocyte activation. In this aim, we will analyze the result of presentation of the full length antigens (AFP, GPC3) and MHC class I-restricted peptide epitopes in HCC patient DC. We will address: CD8+ and CD4+ T cell activation, their phenotype and their function; NK cell activation, and potential for regulatory T cell (Treg) activation or inhibition. A.3 Specific Aim 3. Phase I Pilot Clinical Trial. In this aim, we will enroll and vaccinate 20 AFP+ HCC patients with the multi-antigen loaded DC vaccine. A.4 Specific Aim 4. Immunological Assessment of Vaccine Responses. In this aim, we will characterize immune responses to vaccine-presented antigens; determine NK responses; test for serum tumor markers, biomarkers and Treg and test for correlation with immune and clinical responses. Rational design of vaccines, based on this new information should allow for more rational design of the immune activation to be achieved by newer vaccines.